Hi Alistair:

To obtain observer-free or less-sensor based control, a mathematical model can be employed to obtain some states for feedback control. The reasons are as follows:

(1) Observer-based control need the information (or copy) of nominal controlled system. In this situation, a nominal mathematical model of physical system can obtain the corresponding states. However, the robust stability of the closed-loop system must be considered since a mathematical model is generally inconsistent with true physical system.

(2) The complexity of the closed-loop system is simpler than that using an observer-based control. Hence, the less computation and faster convergent rate of the closed-loop system are accomplished.

The above discussions are for observer-based control experiment. I don't know whether it is your concern. You can refer to our paper and are welcome to discuss more details.

Sincerely,

Chih-Lyang Hwang